Merge pull request #1 from souvik03-136/dev

feat: Add Flask application with Docker deployment

.gitattributes

@@ -0,0 +1,2 @@
+
+* text=auto

.gitignore

@@ -0,0 +1,3 @@
+
+.env
+data

README.md

@@ -15,24 +15,31 @@
 
 
 ## Features
-- [ ]  < feature >
-- [ ]  < feature >
-- [ ]  < feature >
-- [ ]  < feature >
+- [x]  curl -X POST http://127.0.0.1:5000/extract -H "Content-Type: application/json" -d "{\"text\": \"John Doe bought 2 apples for $5\"}"
+
+{"CustomerName":"John Doe","ItemName":"apples","ItemQuantity":"2","Price":"5"}
+
+- [x] curl -X POST "http://127.0.0.1:5000/extract_entities" -H "Content-Type: application/json" -d "{\"text\":\"apples less than 50 rs\"}"
+
+{"action":"less","object":"apples","range":"50"}
+
 
 <br>
 
 ## Dependencies
- - < dependency >
- - < dependency >
+ - Flask==2.3.3
+ - groq==0.9.0
+ - python-dotenv==1.0.1
+ - Werkzeug==2.3.7
 
 
 ## Running
 
 
 < directions to install > 
 ```bash
-< insert code >
+docker-compose up --build 
+docker-compose up 
 ```
 
 < directions to execute >
@@ -46,15 +53,15 @@
 <table>
 	<tr align="center">
 		<td>
-		John Doe
+		Souvik Mahanta
 		<p align="center">
 			<img src = "https://dscvit.com/images/dsc-logo-square.svg" width="150" height="150" alt="Your Name Here (Insert Your Image Link In Src">
 		</p>
 			<p align="center">
-				<a href = "https://github.com/person1">
+				<a href = "https://github.com/souvik03-136">
 					<img src = "http://www.iconninja.com/files/241/825/211/round-collaboration-social-github-code-circle-network-icon.svg" width="36" height = "36" alt="GitHub"/>
 				</a>
-				<a href = "https://www.linkedin.com/in/person1">
+				<a href = "https://www.linkedin.com/in/souvik-mahanta/">
 					<img src = "http://www.iconninja.com/files/863/607/751/network-linkedin-social-connection-circular-circle-media-icon.svg" width="36" height="36" alt="LinkedIn"/>
 				</a>
 			</p>

demand_forecast/Dockerfile

@@ -0,0 +1,20 @@
+# Use an official Python runtime as a parent image
+FROM python:3.9
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Copy the current directory contents into the container at /app
+COPY . /app
+
+# Upgrade pip to the latest version before installing dependencies
+RUN pip install --upgrade pip
+
+# Install necessary dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Expose port 5000 for Flask to run on
+EXPOSE 5000
+
+# Set the default command to run the application
+CMD ["python", "app.py"]

demand_forecast/app.py

@@ -0,0 +1,126 @@
+from flask import Flask, request, jsonify, send_file
+from flask_cors import CORS
+from pathlib import Path
+from forecasting import predict_demand, check_stock_and_alert
+from bokeh_forecast import create_bokeh_plots
+from utils import load_data
+import os
+import logging
+
+app = Flask(__name__)
+
+# Enable CORS
+CORS(app, resources={r"/*": {"origins": "*"}})
+
+# Configure upload folder and file size limit
+UPLOAD_FOLDER = Path(os.getenv('UPLOAD_FOLDER', 'uploads'))
+UPLOAD_FOLDER.mkdir(parents=True, exist_ok=True)
+app.config['MAX_CONTENT_LENGTH'] = 16 * 1024 * 1024  # 16 MB
+
+# Logging setup
+app.logger.setLevel(logging.INFO)
+
+# Allowed file extensions
+ALLOWED_EXTENSIONS = {'csv', 'xlsx'}
+data_file_path = None
+
+
+def allowed_file(filename):
+    """Check if the uploaded file has an allowed extension."""
+    return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
+
+
+@app.route('/')
+def home():
+    return "Welcome to the Bookkeeping AI API! Use /upload to upload a file and /forecast for forecasting."
+
+
+@app.route('/upload', methods=['POST'])
+def upload_file():
+    global data_file_path
+
+    if 'file' not in request.files:
+        return jsonify({"error": "No file part in the request"}), 400
+
+    file = request.files['file']
+    if file.filename == '':
+        return jsonify({"error": "No file selected"}), 400
+
+    if not allowed_file(file.filename):
+        return jsonify({"error": "Only .csv and .xlsx files are allowed"}), 400
+
+    try:
+        # Save the new file and overwrite the existing one
+        data_file_path = UPLOAD_FOLDER / file.filename
+        file.save(data_file_path)
+
+        # Cleanup old plots
+        for plot in UPLOAD_FOLDER.glob("*.html"):
+            plot.unlink()
+
+        app.logger.info(f"File uploaded successfully: {data_file_path}")
+        return jsonify({"message": "File uploaded successfully", "file_path": str(data_file_path)})
+
+    except Exception as e:
+        app.logger.error(f"File upload error: {e}")
+        return jsonify({"error": "Failed to upload file"}), 500
+
+
+@app.route('/forecast', methods=['POST'])
+def forecast():
+    global data_file_path
+    if not data_file_path:
+        return jsonify({"error": "No data file uploaded yet. Please upload a file first."}), 400
+
+    data = request.get_json()
+    item_id = data.get('item_id')
+
+    if item_id is None:
+        return jsonify({"error": "Item ID is required."}), 400
+
+    try:
+        # Load data
+        df = load_data(data_file_path)
+
+        # Validate item ID
+        if item_id not in df['item_id'].values:
+            app.logger.error(f"Item ID {item_id} not found in the dataset")
+            return jsonify({"error": f"Item ID {item_id} not found in the dataset."}), 404
+
+        # Generate predictions
+        future_months, predicted_demand = predict_demand(df, item_id)
+
+        if predicted_demand is None:
+            return jsonify({"error": f"Could not generate forecasts for item ID {item_id}."}), 500
+
+        # Generate alerts and plots
+        alerts = check_stock_and_alert(df, item_id, predicted_demand, future_months)
+        plot_path = create_bokeh_plots(df, item_id, future_months, predicted_demand)
+
+        # Safely convert variables for JSON serialization
+        response = {
+            "future_months": future_months if isinstance(future_months, list) else future_months.tolist(),
+            "predicted_demand": predicted_demand if isinstance(predicted_demand, list) else predicted_demand.tolist(),
+            "alerts": alerts,
+            "plot_url": f"/plot/{item_id}"
+        }
+
+        app.logger.info(f"Forecast successfully generated for item ID {item_id}")
+        return jsonify(response)
+
+    except Exception as e:
+        app.logger.error(f"Error during forecasting: {e}")
+        return jsonify({"error": "An error occurred during forecasting. Please try again."}), 500
+
+
+@app.route('/plot/<item_id>')
+def plot(item_id):
+    plot_path = UPLOAD_FOLDER / f"demand_forecast_{item_id}.html"
+    if plot_path.exists():
+        return send_file(plot_path)
+    else:
+        return jsonify({"error": "Plot not found"}), 404
+
+
+if __name__ == "__main__":
+    app.run(host='0.0.0.0', port=5000, debug=True)

demand_forecast/bokeh_forecast.py

@@ -0,0 +1,91 @@
+from bokeh.plotting import figure, output_file, save
+from bokeh.layouts import column
+from bokeh.models import ColumnDataSource
+import os
+
+
+def create_bokeh_plots(df, item_id, future_months, predicted_demand):
+    # Filter data for the specified item_id
+    item_data = df[df['item_id'] == item_id].copy()
+
+    # Convert transaction date to year-month format
+    item_data['year_month'] = item_data['transaction_date'].dt.to_period('M')
+
+    # Aggregate actual demand data by month
+    actual_demand = item_data.groupby('year_month')['quantity'].sum().reset_index()
+
+    # Prepare ColumnDataSource for actual demand
+    actual_source = ColumnDataSource(data=dict(
+        month=actual_demand['year_month'].dt.to_timestamp(),
+        quantity=actual_demand['quantity']
+    ))
+
+    # Prepare ColumnDataSource for predicted demand
+    predicted_source = ColumnDataSource(data=dict(
+        month=future_months,
+        quantity=predicted_demand
+    ))
+
+    # Create the actual demand plot
+    actual_plot = figure(
+        title=f'Actual Demand for Item ID {item_id}',
+        x_axis_label='Date',
+        y_axis_label='Quantity',
+        x_axis_type='datetime',
+        width=800,
+        height=400,
+        toolbar_location='above',
+        background_fill_color='#f9f9f9'
+    )
+
+    actual_plot.line(
+        'month', 'quantity',
+        source=actual_source,
+        line_width=2,
+        color='blue',
+        legend_label='Actual Demand'
+    )
+
+    actual_plot.scatter(
+        'month', 'quantity',
+        source=actual_source,
+        size=8,
+        color='blue'
+    )
+
+    # Create the predicted demand plot
+    predicted_plot = figure(
+        title=f'Predicted Demand for Item ID {item_id}',
+        x_axis_label='Date',
+        y_axis_label='Quantity',
+        x_axis_type='datetime',
+        width=800,
+        height=400,
+        toolbar_location='above',
+        background_fill_color='#f9f9f9'
+    )
+
+    predicted_plot.line(
+        'month', 'quantity',
+        source=predicted_source,
+        line_width=2,
+        color='orange',
+        legend_label='Predicted Demand'
+    )
+
+    predicted_plot.scatter(
+        'month', 'quantity',
+        source=predicted_source,
+        size=8,
+        color='orange'
+    )
+
+    # Define the HTML file path
+    plot_filename = f"uploads/demand_forecast_{item_id}.html"
+    os.makedirs(os.path.dirname(plot_filename), exist_ok=True)
+
+    # Output the Bokeh plots to an HTML file
+    output_file(plot_filename)
+    save(column(actual_plot, predicted_plot))  # Save the layout to the file instead of showing it
+
+    return plot_filename  # Return the path to the HTML file

demand_forecast/docker-compose.yml

@@ -0,0 +1,10 @@
+version: '3'
+services:
+  app:
+    build: .
+    ports:
+      - "5000:5000"
+    volumes:
+      - .:/app
+    environment:
+      - FLASK_DEBUG=1